Live broadcast bright spot video editing method, device and equipment, and storage medium

Abstract

The invention discloses a live broadcast bright spot video editing method, device and equipment and a storage medium. The method comprises the steps of: obtaining video data and user activity data ofeach anchor user, wherein the user activity data and the video data are synchronous in time, and the user activity data comprises a plurality of user activity dimensions which represent participationbehavior activity degrees of related audience users in the live broadcast process of the anchor users; dividing the video data of each anchor user into a plurality of video clips by taking a preset fixed duration as a unit, and quantitatively determining a comprehensive score of each video clip according to the user activity data to which each user activity dimension belongs; sorting the video clips according to the comprehensive scores, and determining one or more video clips as target clips according to the comprehensive scores; and determining the video data of which the duration span covers the target segment as the bright spot video. According to the embodiment of the invention, related bright spot videos can be identified from historical video data generated by network live broadcastmore accurately and efficiently.

Description

technical field [0001] The present application relates to the technical field of network live broadcasting, and in particular to a method, device, equipment and storage medium for clipping live highlight video. Background technique [0002] In the Internet live broadcast platform, the video data generated by the live broadcast of anchor users can be used for playback, so as to extract some more exciting highlight videos from them, form their personal specials, or forward, share and reuse them for other purposes. [0003] The highlight video is extracted from the video data generated by the live broadcast. The traditional processing method is to manually replay it by the anchor user or the platform operation staff. When the content of the program meets the requirements is found, it is manually edited to finally determine the corresponding highlight video. . [0004] This method of manual target video screening relies on the combination of human subjective feelings and empath...

AI Technical Summary

Problems solved by technology

Although there are various methods, these methods are basically limited to traditional image and / or audio processing technology, and cannot provide an effective means to objectively and systematically determine the real live highlights, that is, the traditional highlight video Editing method, its intelligence is not enough, it is difficult to produce a highlight video that really meets the application requirements

[0007] On the other hand, the video data generated in the live broadcast platform is different from other types of video data. Its content attributes are all-encompassing and cover a wide range. Some known video data mining methods rely solely on image recognition technology or individual parameters in an attempt to determine the highlights of the video. often lead to the determined "highlight video" not being a truly effective highlight segment, and the purpose of accurately determining the real highlight video cannot be achieved

Images